# Prompt-Adapter Context Routing for Parameter-Efficient Multi-Shot Long Video Extrapolation

arXiv: https://arxiv.org/abs/2607.06481 · HTML: https://mlsanity.com/paper/2607.06481/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 53 references that are in the map.
- First posted on arXiv 7 Jul 2026.

## Builds on (most cited first, 27 of 27)

- [Long Context Tuning for Video Generation](https://mlsanity.com/paper/2503.10589/) (2025)
- [Automated Movie Generation via Multi-Agent CoT Planning](https://mlsanity.com/paper/2503.07314/) (2025)
- [The Modern Mathematics of Deep Learning](https://mlsanity.com/paper/2105.04026/) (2021)
- [A Review of Modern Recommender Systems Using Generative Models (Gen-RecSys)](https://mlsanity.com/paper/2404.00579/) (2024)
- [Transferability in Deep Learning: A Survey](https://mlsanity.com/paper/2201.05867/) (2022)
- [Shot2Story: A New Benchmark for Comprehensive Understanding of Multi-shot Videos](https://mlsanity.com/paper/2312.10300/) (2023)
- [StoryMem: Multi-shot Long Video Storytelling with Memory](https://mlsanity.com/paper/2512.19539/) (2025)
- [Deep learning observables in computational fluid dynamics](https://mlsanity.com/paper/1903.03040/) (2019)
- [VideoGen-of-Thought: Step-by-step generating multi-shot video with minimal manual intervention](https://mlsanity.com/paper/2503.15138/) (2025)
- [Learn to Accumulate Evidence from All Training Samples: Theory and Practice](https://mlsanity.com/paper/2306.11113/) (2023)
- [ShotStream: Streaming Multi-Shot Video Generation for Interactive Storytelling](https://mlsanity.com/paper/2603.25746/) (2026)
- [CoS: Chain-of-Shot Prompting for Long Video Understanding](https://mlsanity.com/paper/2502.06428/) (2025)
- [One-Shot Learning of Multi-Step Tasks from Observation via Activity Localization in Auxiliary Video](https://mlsanity.com/paper/1806.11244/) (2018)
- [Assessing Visual Quality of Omnidirectional Videos](https://mlsanity.com/paper/1709.06342/) (2017)
- [A2RD: Agentic Autoregressive Diffusion for Long Video Consistency](https://mlsanity.com/paper/2605.06924/) (2026)
- [ReCA: Multi-Shot Long Video Extrapolation via Recursive Context Allocation](https://mlsanity.com/paper/2605.26525/) (2026)
- [FocusChat: Text-guided Long Video Understanding via Spatiotemporal Information Filtering](https://mlsanity.com/paper/2412.12833/) (2024)
- [Not All Frames Deserve Full Computation: Accelerating Autoregressive Video Generation via Selective Computation and Predictive Extrapolation](https://mlsanity.com/paper/2604.02979/) (2026)
- [Soap2Soap: Long Cinematic Video Remaking via Multi-Agent Collaboration](https://mlsanity.com/paper/2605.17423/) (2026)
- [GroundShot: Visually Consistent Multi-Shot Long Video Generation via Entity-Grounded Shot Scheduling](https://mlsanity.com/paper/2606.20799/) (2026)
- [DILIE: Deep Internal Learning for Image Enhancement](https://mlsanity.com/paper/2012.06469/) (2020)
- [Super Encoding Network: Recursive Association of Multi-Modal Encoders for Video Understanding](https://mlsanity.com/paper/2506.07576/) (2025)
- [A multitask deep learning model for real-time deployment in embedded systems](https://mlsanity.com/paper/1711.00146/) (2017)
- [Monodense Deep Neural Model for Determining Item Price Elasticity](https://mlsanity.com/paper/2603.29261/) (2026)
- [Bridging Your Imagination with Audio-Video Generation via a Unified Director](https://mlsanity.com/paper/2512.23222/) (2025)
- [DQ-Ladder: A Deep Reinforcement Learning-based Bitrate Ladder for Adaptive Video Streaming](https://mlsanity.com/paper/2603.13597/) (2026)
- [HiCrew: Hierarchical Reasoning for Long-Form Video Understanding via Question-Aware Multi-Agent Collaboration](https://mlsanity.com/paper/2604.21444/) (2026)

## Abstract (arXiv)

We present PACR-Video, a parameter-efficient framework for multi-shot long video extrapolation that preserves recurring entities, scene structure, visual style, and causal progression without full generator fine-tuning. PACR-Video keeps a text-to-video diffusion transformer frozen and augments it with low-rank temporal adapters conditioned by learned shot-role prompt tokens. To maintain long-horizon coherence, it builds a recursive prompt bank that stores compact entity, location, action, and style prompts from previous shots, then routes them through adapter gates according to predicted narrative dependencies. A Shot-Local/Story-Global tuning objective combines next-shot reconstruction, cross-shot identity contrast, and prompt sparsity regularization, while an adapter composition schedule balances early-shot visual consistency with later-shot event progression and viewpoint change. Across six multi-shot and long-video benchmarks, PACR-Video outperforms text-to-video, tuning-based, memory-augmented, streaming, and recursive-context baselines on distributional quality, semantic alignment, identity consistency, temporal smoothness, motion stability, transition coherence, and human preference. These results show that compact prompt routing and lightweight temporal adaptation provide sufficient controllable capacity for stable long video extrapolation.